Department of Internal Medicine, VieCuri Medical Centre, Venlo, The Netherlands
Lucas A Ramos
Department of Biomedical Engineering and Physics/Department of Epidemiology & Data Science, Amsterdam University Medical Centres, Duivendrecht, The Netherlands
Paul W G Elbers
Department of Intensive Care Medicine, Amsterdam Medical Data Science, Amsterdam Cardiovascular Science, Amsterdam Infection and Immunity Institute, Amsterdam UMC Location VUmc, Amsterdam, The Netherlands
W Joost Wiersinga
Center for Experimental and Molecular Medicine (C.E.M.M.), Amsterdam UMC Locatie AMC, Amsterdam, The Netherlands
Maarten C Ottenhoff
Department of Neurosurgery, Maastricht University, Maastricht, The Netherlands
Wouter Potters
Department of Neurology, Amsterdam University Medical Centres, Duivendrecht, The Netherlands
Deborah Hubers
Department of Intensive Care, Maastricht Universitair Medisch Centrum+, Maastricht, The Netherlands
Shi Hu
Informatics Institute, University of Amsterdam, Amsterdam, The Netherlands
Egill A Fridgeirsson
Department of Psychiatry, Amsterdam University Medical Centres, Duivendrecht, The Netherlands
Rajat Thomas
Department of Psychiatry, Amsterdam University Medical Centres, Duivendrecht, The Netherlands
Christian Herff
Department of Neurosurgery, Maastricht University, Maastricht, The Netherlands
Pieter Kubben
Department of Neurosurgery, Maastricht Universitair Medisch Centrum+, Maastricht, The Netherlands
Max Welling
Informatics Institute, University of Amsterdam, Amsterdam, The Netherlands
Lucas M Fleuren
Department of Intensive Care, Amsterdam University Medical Centres, Duivendrecht, Noord-Holland, The Netherlands
Michiel Schinkel
1 Department of Internal Medicine, Division of Acute Medicine, Amsterdam UMC Locatie VUmc, Amsterdam, The Netherlands
Peter G Noordzij
1 Department of Anaesthesiology, Intensive Care and Pain management, St. Antonius Hospital, Nieuwegein, Netherlands
Caroline E Wyers
Department of Internal Medicine, VieCuri Medical Centre, Venlo, The Netherlands
David T B Buis
Department of Internal Medicine, Amsterdam UMC Locatie VUmc, Amsterdam, The Netherlands
Ella H C van den Hout
Department of Internal Medicine, Noordwest Ziekenhuisgroep, Alkmaar, The Netherlands
Daisy Rusch
Research, Martini Ziekenhuis, Groningen, Netherlands
Kim C E Sigaloff
Infectious Diseases and Amsterdam Institute for Infection and Immunology, Amsterdam UMC, Vrije Universiteit, Amsterdam, The Netherlands
Renee A Douma
resident internal medicine
Lianne de Haan
Department of Internal Medicine, Flevo Hospital, Almere, The Netherlands
Niels C Gritters van den Oever
Department of Intensive Care, Treant Zorggroep, Hoogeveen, Netherlands
Guido A van Wingen
Department of Psychiatry, University of Amsterdam, Amsterdam, The Netherlands
Objective Develop and validate models that predict mortality of patients diagnosed with COVID-19 admitted to the hospital.Design Retrospective cohort study.Setting A multicentre cohort across 10 Dutch hospitals including patients from 27 February to 8 June 2020.Participants SARS-CoV-2 positive patients (age ≥18) admitted to the hospital.Main outcome measures 21-day all-cause mortality evaluated by the area under the receiver operator curve (AUC), sensitivity, specificity, positive predictive value and negative predictive value. The predictive value of age was explored by comparison with age-based rules used in practice and by excluding age from the analysis.Results 2273 patients were included, of whom 516 had died or discharged to palliative care within 21 days after admission. Five feature sets, including premorbid, clinical presentation and laboratory and radiology values, were derived from 80 features. Additionally, an Analysis of Variance (ANOVA)-based data-driven feature selection selected the 10 features with the highest F values: age, number of home medications, urea nitrogen, lactate dehydrogenase, albumin, oxygen saturation (%), oxygen saturation is measured on room air, oxygen saturation is measured on oxygen therapy, blood gas pH and history of chronic cardiac disease. A linear logistic regression and non-linear tree-based gradient boosting algorithm fitted the data with an AUC of 0.81 (95% CI 0.77 to 0.85) and 0.82 (0.79 to 0.85), respectively, using the 10 selected features. Both models outperformed age-based decision rules used in practice (AUC of 0.69, 0.65 to 0.74 for age >70). Furthermore, performance remained stable when excluding age as predictor (AUC of 0.78, 0.75 to 0.81).Conclusion Both models showed good performance and had better test characteristics than age-based decision rules, using 10 admission features readily available in Dutch hospitals. The models hold promise to aid decision-making during a hospital bed shortage.
